航空貨運承攬業併裝決策模式之研究

Abstract

隨著油價的持續上漲及市場上的價格競爭，台灣的航空貨運承攬業者在短短的十年間，從高收益的行業變的愈來愈薄利化。除了不斷地朝多元化發展以創造收益來源外，減少在成本上的支出亦成為獲利的重要方式。而併裝利潤是空運承攬業者的主要收益來源，目前市場上航空貨運承攬業者在併貨時大多是由公司內「併裝部門」的人員自行依經驗決定如何併貨。而這些在實務作業上所發展出來的併裝決策方式是否確能帶給業者最大的獲利? 所為之併裝決策是否尚有改善的空間? 似乎較少業者會針對此部份來進行評估或檢討。

航空運費計價的特性有兩項：一是「同時考慮重量與體積」；一是「數量折扣」- 即對於不同之重量有不同之費率級距(重量愈重，費率愈低)。航空貨運承攬業基於此兩項特性，透過併裝承攬貨物的方式來降低支付給航空公司的總運價。但由於空運所要處理的貨物多對時效有極高的要求，業者需在極短的時間內完成併貨、繕打文件、報關等手續：而併裝人員亦多需於極短的時間內作出判斷如何併貨。且由於實務上來自客戶之訂位均為動態、持續且具不確定性的，為了讓後續單位有足夠時間完成相關作業，併裝人員無法等到所有貨物資料都有了之後再決定如何併裝，而需於陸續接受訂位時就一邊陸續併貨；如此一來大幅增加了該併裝決策之困難度。

本研究深入探討航空貨運承攬業在併裝實務上的作業，剖析目前業者在安排併裝時所為之決策過程、並將其模式化，進一步嘗試透過MatLab的程式撰寫來模擬其動態決策過程、並進行數值測試。且為求數值測試時所採用之貨物資料能與實際業者出貨情況儘量一致，本研究統計業者實際出貨資料，並發展出一套亂數模擬之方式來產生此種可同時考量兩種屬性(重量及密度)且屬性相關之亂數資料。進而對所建模式來進行敏感度分析及情境分析，以瞭解影響併裝結果的因素。

此外，為比較經此種動態併裝決策模式調整過之併裝是否尚有改善的空間，本研究亦將導入文獻中針對併裝問題的「數量折扣」及「同時考量重量及體積」等特色所設計出的静態數學規劃求解模式(透過LINGO作運算)，以比較不同模式下併裝結果之差異。

With the soaring oil price and the competition in the market, the air cargo forwarders faced the change from a high revenue industry to a discouraging low profit business in the past 10 years. Beside the constant need to diversify the service so as to create the revenue and not to fall in the price competition, saving the cost on the expenditure to airlines is also an important way to increase the profit. The consolidation profit is the main profit source for air cargo forwarders. But at present, the consolidation is mostly done by the consolidation department of each forwarder; the operator consolidates the freight by personal experience and skill. Whether those consolidation decision skills developed through the daily practice can actually bring the most possible profit margin to forwarders? Is there any room for improvement? It seems rarely any forwarder spent time on this issue.

The 2 key features on air freight pricing: one is the so called “Chargeable weight” and the other is “Weight discount” (the heavier the weight break is; the lower its air freight rate will be). Basing on these 2 features, the freight forwarders will usually try to save the expenditure they pay to airlines by “smart consolidation”. But for most cargoes shipped by air, the customers are usually very demanding on the lead time. Forwarders need to finish the operations including freight consolidation, document making and customs declaration…etc in very short time. The consolidation operators usually have to judge how to consolidate the cargoes in very short time. And because the booking from customer comes irregularly and continuously (or dynamically), if the operators wait for all cargo detail and then make the consolidation, there’ll be no time for the operations afterwards. To consolidate the cargoes while receiving the booking makes the consolidation decision much more difficult.

This research studies in detail the consolidation operations in real forwarding world; analyze the decision process step by step and then modelize it. A further numerical testing was taken by using MatLab to write the program for simulating the dynamic consolidation decision process. And in order to simulate the cargo booking a forwarder may face in a real world, this research collected the actual booking detail for some period of time from the forwarder and developed a random number generating method to generate the cargo data (by using MatLab random number generator) which can simulate the 2 attributes of actual cargoes at the same time (the weight and the density). Through the numerical testing, we got a clear picture for the factors and conditions which have the key influence on the consolidation result.

Besides, to understand whether there’s still room for improvement on the consolidation done by skilled operators, the mathematical model for solving static consolidation questions (making use of LINGO optimization solver) will also be adopted for the comparison of consolidation results by different models.